XGBoost wrapper

XGBoost wrapper for seurat project in R

source("tianfengRwrappers.R")

载入需要的程辑包：dplyr

载入程辑包：‘dplyr’

The following objects are masked from ‘package:stats’:

    filter, lag

The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union

载入需要的程辑包：reticulate
载入需要的程辑包：tidyr

载入程辑包：‘MySeuratWrappers’

The following objects are masked from ‘package:Seurat’:

    DimPlot, DoHeatmap, LabelClusters, RidgePlot, VlnPlot


载入程辑包：‘cowplot’

The following object is masked from ‘package:ggpubr’:

    get_legend

载入需要的程辑包：viridisLite

载入程辑包：‘reshape2’

The following object is masked from ‘package:tidyr’:

    smiths

NOTE: Either Arial Narrow or Roboto Condensed fonts are required to use these themes.
      Please use hrbrthemes::import_roboto_condensed() to install Roboto Condensed and
      if Arial Narrow is not on your system, please see https://bit.ly/arialnarrow

Registered S3 method overwritten by 'enrichplot':
  method               from
  fortify.enrichResult DOSE
clusterProfiler v3.14.3  For help: https://guangchuangyu.github.io/software/clusterProfiler

If you use clusterProfiler in published research, please cite:
Guangchuang Yu, Li-Gen Wang, Yanyan Han, Qing-Yu He. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS: A Journal of Integrative Biology. 2012, 16(5):284-287.
Registering fonts with R

载入程辑包：‘plotly’

The following object is masked from ‘package:ggplot2’:

    last_plot

The following object is masked from ‘package:stats’:

    filter

The following object is masked from ‘package:graphics’:

    layout

载入需要的程辑包：Biobase
载入需要的程辑包：BiocGenerics
载入需要的程辑包：parallel

载入程辑包：‘BiocGenerics’

The following objects are masked from ‘package:parallel’:

    clusterApply, clusterApplyLB, clusterCall, clusterEvalQ, clusterExport, clusterMap, parApply,
    parCapply, parLapply, parLapplyLB, parRapply, parSapply, parSapplyLB

The following objects are masked from ‘package:dplyr’:

    combine, intersect, setdiff, union

The following objects are masked from ‘package:stats’:

    IQR, mad, sd, var, xtabs

The following objects are masked from ‘package:base’:

    anyDuplicated, append, as.data.frame, basename, cbind, colnames, dirname, do.call, duplicated,
    eval, evalq, Filter, Find, get, grep, grepl, intersect, is.unsorted, lapply, Map, mapply, match,
    mget, order, paste, pmax, pmax.int, pmin, pmin.int, Position, rank, rbind, Reduce, rownames,
    sapply, setdiff, sort, table, tapply, union, unique, unsplit, which, which.max, which.min

Welcome to Bioconductor

    Vignettes contain introductory material; view with 'browseVignettes()'. To cite Bioconductor, see
    'citation("Biobase")', and for packages 'citation("pkgname")'.

载入需要的程辑包：e1071

载入程辑包：‘widgetTools’

The following object is masked from ‘package:dplyr’:

    funs


载入程辑包：‘DynDoc’

The following object is masked from ‘package:BiocGenerics’:

    path


载入程辑包：‘DT’

The following object is masked from ‘package:Seurat’:

    JS

========================================
circlize version 0.4.13
CRAN page: https://cran.r-project.org/package=circlize
Github page: https://github.com/jokergoo/circlize
Documentation: https://jokergoo.github.io/circlize_book/book/

If you use it in published research, please cite:
Gu, Z. circlize implements and enhances circular visualization
  in R. Bioinformatics 2014.

This message can be suppressed by:
  suppressPackageStartupMessages(library(circlize))
========================================

载入需要的程辑包：grid
========================================
ComplexHeatmap version 2.2.0
Bioconductor page: http://bioconductor.org/packages/ComplexHeatmap/
Github page: https://github.com/jokergoo/ComplexHeatmap
Documentation: http://jokergoo.github.io/ComplexHeatmap-reference

If you use it in published research, please cite:
Gu, Z. Complex heatmaps reveal patterns and correlations in multidimensional 
  genomic data. Bioinformatics 2016.
========================================


载入程辑包：‘ComplexHeatmap’

The following object is masked from ‘package:plotly’:

    add_heatmap

library(xgboost)

载入程辑包：‘xgboost’

The following object is masked from ‘package:plotly’:

    slice

The following object is masked from ‘package:dplyr’:

    slice

library(Matrix)

载入程辑包：‘Matrix’

The following objects are masked from ‘package:tidyr’:

    expand, pack, unpack

library(mclust)

    __  ___________    __  _____________
   /  |/  / ____/ /   / / / / ___/_  __/
  / /|_/ / /   / /   / / / /\__ \ / /   
 / /  / / /___/ /___/ /_/ /___/ // /    
/_/  /_/\____/_____/\____//____//_/    version 5.4.9
Type 'citation("mclust")' for citing this R package in publications.

library(tidyverse)

Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
Registered S3 method overwritten by 'cli':
  method     from    
  print.boxx spatstat
─ Attaching packages ───────────────────────────────────── tidyverse 1.3.1 ─
✓ tibble  3.1.5     ✓ stringr 1.4.0
✓ readr   2.1.2     ✓ forcats 0.5.1
✓ purrr   0.3.4     
─ Conflicts ─────────────────────────────────────── tidyverse_conflicts() ─
x Biobase::combine()       masks BiocGenerics::combine(), dplyr::combine()
x Matrix::expand()         masks tidyr::expand()
x plotly::filter()         masks dplyr::filter(), stats::filter()
x widgetTools::funs()      masks dplyr::funs()
x dplyr::lag()             masks stats::lag()
x purrr::map()             masks mclust::map()
x Matrix::pack()           masks tidyr::pack()
x BiocGenerics::Position() masks ggplot2::Position(), base::Position()
x purrr::simplify()        masks clusterProfiler::simplify()
x xgboost::slice()         masks plotly::slice(), dplyr::slice()
x Matrix::unpack()         masks tidyr::unpack()

library(SHAPforxgboost)
library(lambda.r)

载入程辑包：‘lambda.r’

The following object is masked from ‘package:reticulate’:

    %as%

ds0 <- readRDS("ds0.rds")
ds1 <- readRDS("ds1.rds")
ds2 <- readRDS("ds2.rds")

XGBoost_predict_from_seuobj <- function(seuobj, bst_model, is_highvar = T, seed = 7, celltype_assign = 2)
{
  #return a updated seurat object with new metadata named confidence and projected_idents
  seuobj_label <- as.numeric(as.character(Idents(seuobj)))
  if(is.na(seuobj_label[1])) # check vaild Idents
  {
    stop("Please ensure that seurat idents are in numeric forms")
  }
  temp <- get_data_table(seuobj, highvar = T, type = "data")
  seuobj_data <- matrix(data = 0, nrow = bst_model$nfeatures, ncol = length(colnames(temp)), 
                     byrow = FALSE, dimnames = list(bst_model[["feature_names"]],colnames(temp)))
  intersect_features <- intersect(bst_model[["feature_names"]], rownames(temp))
  seuobj_data[intersect_features,] <- temp[intersect_features,]
  rm(temp)
  

  # colnames(seuobj_data) <- NULL
  seuobj_test_data <- list(data = t(as(seuobj_data,"dgCMatrix")), label = seuobj_label)
  seuobj_test <- xgb.DMatrix(data = seuobj_test_data$data,label = seuobj_test_data$label)
  
  #预测结果
  predict_seuobj_test <- predict(bst_model, newdata = seuobj_test)
  
  predict_prop_seuobj <<- matrix(data=predict_seuobj_test, nrow = bst_model[["params"]][["num_class"]], 
                             ncol = ncol(seuobj), byrow = FALSE, 
                             dimnames = list(as.character(0:(bst_model[["params"]][["num_class"]]-1)),
                                             colnames(seuobj)))

  # predict cell types
  if(celltype_assign = 1){

错误: 意外的'=' in:
"  # predict cell types
  if(celltype_assign ="

---

function instance

train

function instance

predict

Idents(ds0) <- ds0$seurat_clusters
ds0 <- XGBoost_predict_from_seuobj(ds0, bst_model = bst_model) %>% 
  project2ref_celltype(ref_seuobj = ds2, ref_labels = c("seurat_clusters","Classification1"))

ds1 <- XGBoost_predict_from_seuobj(ds1, bst_model = bst_model) %>% 
  project2ref_celltype(ref_seuobj = ds2, ref_labels = c("seurat_clusters","Classification1"))
umapplot(ds0, group.by = "ref_celltype")

scmap wrapper

library(SingleCellExperiment)
library(scmap)

mkref_scmap_from_seuobj <- function(ref_seuobj){
  # return a ref sce object
  ref_sce <- as.SingleCellExperiment(ref_seuobj)
  logcounts(ref_sce) <- log2(counts(ref_sce) + 1)
  
  counts(ref_sce) <- as.matrix(counts(ref_sce))
  logcounts(ref_sce) <- as.matrix(logcounts(ref_sce))
  
  rowData(ref_sce)$feature_symbol <- rownames(ref_sce)
  ref_sce <- ref_sce[!duplicated(rownames(ref_sce)), ]
  ref_sce <- selectFeatures(ref_sce, suppress_plot = T) %>% indexCell()

  return(ref_sce)
}

query_scmap_from_refsce <- function(query_seuobj, ref_sce, ref_labels = 'Classification1'){
  #return a updated seurat object with new metadata named scmap_idents
  
  query_sce <- as.SingleCellExperiment(query_seuobj)
  logcounts(query_sce) <- log2(counts(query_sce) + 1)
  
  counts(query_sce) <- as.matrix(counts(query_sce))
  logcounts(query_sce) <- as.matrix(logcounts(query_sce))
  
  rowData(query_sce)$feature_symbol <- rownames(query_sce)
  query_sce <- query_sce[!duplicated(rownames(query_sce)), ]
  query_sce <- selectFeatures(query_sce, suppress_plot = T) %>% indexCell()
  
  scmapCell_results <- scmapCell(query_sce, list(ref = metadata(ref_sce)$scmap_cell_index))
  scmapCell_clusters <- scmapCell2Cluster(scmapCell_results,
                                          list(as.character(colData(ref_sce)[[ref_labels]])))
  
  query_seuobj$scmap_idents <- data.frame(scmapCell_clusters$scmap_cluster_labs[,"ref"],
                                        row.names = colnames(query_seuobj))
  return(query_seuobj)
}

ds1 <- query_scmap_from_refsce(ds1, ref_sce)

Warning: 'isSpike' is deprecated.
See help("Deprecated")
Parameter M was not provided, will use M = n_features / 10 (if n_features <= 1000), where n_features is the number of selected features, and M = 100 otherwise.
Parameter k was not provided, will use k = sqrt(number_of_cells)
Warning: stack imbalance in '<-', 2 then 1

supervised vs unsupervised clustering

upset plot

SMC2

# library(UpSetR)

df <- cbind(ds1[["Classification1"]],ds1[["ref_celltype"]],ds1[["scmap_idents"]])
df <- df[df$Classification1 == "SMC2" | df$ref_celltype == "SMC2" | df$scmap_idents == "SMC2" ,]
li <- table(df) %>% as.data.frame() #获得含有SMC2的frequency
li <- li[!(li$Freq < 5),] #删除frequency<5的行
li <- li[order(li$Freq,decreasing = T),]
dd <- data.frame(li,index = as.character(1:nrow(li)))

dd$index <- factor(dd$index,levels = 1:length(levels(dd$index)))
dd$Freq <- NULL
dd <- reshape2::melt(dd,id.var = "index")
colnames(dd) <- c("index","type","name")

dt <- data.frame(Freq = li$Freq,index = as.character(1:nrow(li)))
dt$index <- factor(dt$index,levels = 1:length(levels(dt$index)))
dt$col <- dt$Freq>(sum(dt$Freq)/10) #set the color to red for those frequency > 0.1

p1 <- ggplot(dd)+geom_point(mapping = aes(x = index, y = type, color = name), size = 6) + mytheme2 + scale_color_manual(values = colors_list) + theme(axis.ticks.x = element_blank(),axis.title.x = element_blank())
p2 <- ggplot(dt,aes(x = index, y = Freq, fill = col))+geom_bar(stat = "identity") + mytheme2 + theme(axis.ticks.x = element_blank(),axis.title.x = element_blank(), legend.position = 'none') + scale_fill_manual(values = c("black","red"))
 
plot <- cowplot::plot_grid(p2,p1,ncol = 1,align = 'v',rel_heights = c(2,1))
plot
# ggsave("SMC2cells.svg",plot=plot,device = svg,height = 8,width = 8)

SMC1

df <- cbind(ds1[["Classification1"]],ds1[["ref_celltype"]],ds1[["scmap_idents"]])
df <- df[df$Classification1 == "SMC1" | df$ref_celltype == "SMC1" | df$scmap_idents == "SMC1" ,]
li <- table(df) %>% as.data.frame() #获得含有SMC1的frequency
li <- li[!(li$Freq < 5),] #删除frequency<5的行
li <- li[order(li$Freq,decreasing = T),]
dd <- data.frame(li,index = as.character(1:nrow(li)))

dd$index <- factor(dd$index,levels = 1:length(levels(dd$index)))
dd$Freq <- NULL
dd <- reshape2::melt(dd,id.var = "index")
colnames(dd) <- c("index","type","name")

dt <- data.frame(Freq = li$Freq,index = as.character(1:nrow(li)))
dt$index <- factor(dt$index,levels = 1:length(levels(dt$index)))
dt$col <- dt$Freq>(sum(dt$Freq)/10) #set the color to red for those frequency > 0.1

p1 <- ggplot(dd)+geom_point(mapping = aes(x = index, y = type, color = name), size = 6) + mytheme2 + scale_color_manual(values = colors_list) + theme(axis.ticks.x = element_blank(),axis.title.x = element_blank())
p2 <- ggplot(dt,aes(x = index, y = Freq, fill = col))+geom_bar(stat = "identity") + mytheme2 + theme(axis.ticks.x = element_blank(),axis.title.x = element_blank(), legend.position = 'none') + scale_fill_manual(values = c("black","red"))
 
plot <- cowplot::plot_grid(p2,p1,ncol = 1,align = 'v',rel_heights = c(2,1))
plot
ggsave("SMC1cells.svg",plot=plot,device = svg,height = 8,width = 8)

FBM

df <- cbind(ds1[["Classification1"]],ds1[["ref_celltype"]],ds1[["scmap_idents"]])
df <- df[df$Classification1 == "Fibromyocyte" | df$ref_celltype == "Fibromyocyte" | df$scmap_idents == "Fibromyocyte" ,]
li <- table(df) %>% as.data.frame() #获得含有Fibromyocyte的frequency
li <- li[!(li$Freq < 5),] #删除frequency<5的行
li <- li[order(li$Freq,decreasing = T),]
dd <- data.frame(li,index = as.character(1:nrow(li)))

dd$index <- factor(dd$index,levels = 1:length(levels(dd$index)))
dd$Freq <- NULL
dd <- reshape2::melt(dd,id.var = "index")
colnames(dd) <- c("index","type","name")

dt <- data.frame(Freq = li$Freq,index = as.character(1:nrow(li)))
dt$index <- factor(dt$index,levels = 1:length(levels(dt$index)))
dt$col <- dt$Freq>(sum(dt$Freq)/10) #set the color to red for those frequency > 0.1

p1 <- ggplot(dd)+geom_point(mapping = aes(x = index, y = type, color = name), size = 6) + mytheme2 + scale_color_manual(values = colors_list) + theme(axis.ticks.x = element_blank(),axis.title.x = element_blank())
p2 <- ggplot(dt,aes(x = index, y = Freq, fill = col))+geom_bar(stat = "identity") + mytheme2 + theme(axis.ticks.x = element_blank(),axis.title.x = element_blank(), legend.position = 'none') + scale_fill_manual(values = c("black","red"))
 
plot <- cowplot::plot_grid(p2,p1,ncol = 1,align = 'v',rel_heights = c(2,1))
plot
ggsave("Fibromyocytecells.svg",plot=plot,device = svg,height = 8,width = 8)

correlation plot

PCA

unsupervised version

ds1 <- XGBoost_predict_from_seuobj(ds1,bst_model)

Warning in XGBoost_predict_from_seuobj(ds1, bst_model) :
  Please ensure that seurat idents are in numeric forms

[1] "ARI = 0.306022542455434"
[1] "return a seurat object with meta.data'X1'~'Xn'"
[1] "return a seurat object with meta.data'projected_idents'"

ds1 <- XGBoost_predict_from_seuobj(ds1,bst_model)

Warning in XGBoost_predict_from_seuobj(ds1, bst_model) :
  Please ensure that seurat idents are in numeric forms

[1] "ARI = 0.306022542455434"
[1] "return a seurat object with meta.data'X1'~'Xn'"
[1] "return a seurat object with meta.data'projected_idents'"

ds2 <- XGBoost_predict_from_seuobj(ds2,bst_model)

Warning in XGBoost_predict_from_seuobj(ds2, bst_model) :
  强制改变过程中产生了NA
Warning in XGBoost_predict_from_seuobj(ds2, bst_model) :
  Please ensure that seurat idents are in numeric forms

[1] "ARI = 0.744493360267919"
[1] "return a seurat object with meta.data'X1'~'Xn'"
[1] "return a seurat object with meta.data'projected_idents'"

umapplot(ds0,"projected_idents") /
umapplot(ds1,"projected_idents") /
umapplot(ds2,"projected_idents")

Scale for 'colour' is already present. Adding another scale for 'colour', which will replace the
existing scale.
Scale for 'colour' is already present. Adding another scale for 'colour', which will replace the
existing scale.
Scale for 'colour' is already present. Adding another scale for 'colour', which will replace the
existing scale.

# mergeds <- mergeds %>% SCTransform(vars.to.regress = "percent.mt", verbose = F) %>% 
#     RunPCA() %>% FindNeighbors(dims = 1:20) %>% 
#     RunUMAP(dims = 1:20) %>% 
#     FindClusters(resolution = 0.1)
# 
# umapplot(mergeds,group.by = "Classification1" ,split.by = "orig.ident")

Add a new chunk by clicking the Insert Chunk button on the toolbar or by pressing Ctrl+Alt+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Ctrl+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.